OpenAI’s Head of ChatGPT Nick Turley and Chief Research Officer Mark Chen took a look at ChatGPT on the OpenAI podcast released yesterday.
From:
- GPT origin stories to
- Plans for agentic coding to
- The importance of feedback
– these topics and more were covered.
Early in the podcast [13:00], podcast host Andrew Mayne, a former OpenAI employee, briefly touched on ads and posed a question to the OpenAI executives about managing the value tradeoff between ChatGPT and its users.
(Transcript lightly edited.)
OpenAI podcast host Andrew Mayne:
“…the problem with social media is you’re basically monetized by engagement time. You want to keep people on there longer, so you can show them more ads.
And certainly, the more people use ChatGPT, obviously, there’s a cost to OpenAI. The idea is… use it once and stay around forever, but that’s not practical.”
“How do you weigh that?… i.e. the idea of making people happy with what they’re getting versus making the model be broadly more useful than just pleasing.”
ChatGPT’s Nick Turley replied:
“I feel very lucky in this regard because we have a product that’s very utilitarian. People use it to either achieve things that they do know how to do, but don’t feel like doing faster or with less effort, or they’re using it to do things that they couldn’t do at all.
The first example is writing an email that you’ve been dreading.
The second example might be running a data analysis that you didn’t actually know how to do in Excel.
So, those are very utilitarian things. And fundamentally, as you improve, you have to spend less time on the product, right? Ideally, it takes less turns back and forth, or maybe you actually delegate to the AI, so you’re not in the product at all.”
So for us, ‘time spent,’ – it’s very much not the thing we optimize for.
We do care about your long-term retention, because we do think that’s a sign of value. If you’re coming back three months later, that clearly means we did something right. But what that means is, you know, I always say, ‘Show me the incentive, and I’ll show you the outcome.’
We have, I think, the right fundamental incentives to build something great. That doesn’t mean we’ll always get it right. The sycophancy events (in April) were important and good learning for us, and I’m proud of how we acted on it.
But fundamentally, I think we have the right setup to build something awesome.”
Hear more on the Apple Podcasts app. (1 hour, 7 minute duration)
More: Did you know GPT stands for “Generative pretrained transformer”… a type of LLM? You do now.
Traffic referral (non-bot) anecdote
In a post on LinkedIn yesterday, casual games developer Arkadium CEO Kenny Rosenblatt discussed traffic to his company’s online games websites:
It may be only one anecdote, but this graphic shows ChatGPT’s dominance.
New publisher revenue stream?
Cloudflare CEO Matthew Prince’s news-making appearance at Cannes, where he decried the huge increase of website crawling by AI chatbots, quickly became a rallying cry for publishers.
Two weeks later, Cloudflare announced a new potential revenue stream for publishers.
Read the deluge of announcements from the company yesterday:
- “Cloudflare Just Changed How AI Crawlers Scrape the Internet-at-Large; Permission-Based Approach Makes Way for A New Business Model” (July 1) – Cloudflare press release
- “Content Independence Day: no AI crawl without compensation!” (July 1) – Matthew Prince on the Cloudflare blog
- “Control content use for AI training with Cloudflare’s managed robots.txt and blocking for monetized content” (July 1) – Cloudflare blog
Computerworld summarized the news:
“Cloudflare will block AI bots from crawling websites by default for new customers, and broker pay-per-crawl deals between its customers and bot operators…
Cloudflare will block AI crawlers from accessing new customers’ websites without permission starting July 1 and is testing a way to make AI pay for the data it gathers”
New publisher revenue stream – reaction
On LinkedIn, Raptive Chief Strategy Officer Paul Bannister of ad tech firm Raptive said in part regarding the Cloudflare news:
“We’ve been a part of the group of publishers and media companies discussing this with Cloudflare over the last few months. We’re really excited to see it go live and start seeing the impact that it causes.
Of course, Google is exempted from all of this. By using their monopoly position to tie together their search bot and their AI bot, they make it impossible for anyone to block their AI scrapers without also losing their search traffic.
While there have been some companies around for a while trying to do this, I think Cloudflare’s announcement is the start of a new generation of ideas that actually will start to turn the tide in favor of content creators. More good things are coming and it’s an exciting time!”
In the comments to Bannister’s post, IAB Tech Lab CEO Anthony Katsur chimed in:
Don’t forget the IAB Tech Lab‘s recently launched LLM Content Ingest API initiative, which aims to standardize efforts around these efforts with publishers and edge providers.
https://iabtechlab.com/standards/llm-framework/
Firsthand co-founder Michael Rubenstein wrote on LinkedIn yesterday:
“Today’s announcement by Cloudflare is a meaningful step forward for publishers and toward a more balanced AI economy.
By blocking AI crawlers by default, Cloudflare is helping publishers reclaim control over how their content is accessed and monetized — and pushing the ecosystem toward a model that values the work of content creators…”
New publisher revenue stream? – analysis
Enter the bot “ad” network, where the “ads” are publishers’ website content. Cloudflare, an internet infrastructure management company, will now be a broker. Nice margins if you can get ‘em?
The true extent of the opportunity for publishers in charging AI crawlers on a per crawl basis will be interesting to watch.
Among the features of the new Cloudflare product: publishers can guide crawlers to content that does NOT have ads if they so desire.
It’s hard to imagine that revenue will be significant the further out one goes out on the long tail of publishers. Of course, with Cloudflare “brokering the deal” by aggregating the long tail, revenue could be impactful for Cloudflare’s bottom line.
For savvy, large (highly trafficked) publishers, bot crawling deals would likely be direct rather than going through Cloudflare.
What’s old is new again – hello, bot “ad” network.
TechCrunch makes the case on why Cloudflare is uniquely positioned for this opportunity. Read it.
Without question this is also great marketing by Cloudflare to promote its sell-side, website performance tools.
Agencies AI downgrade
After visiting the recent Cannes Lions ad festival, Barclays financial analysts said in a note to investors earlier this week that they were even more bearish about agency holding companies in light of the challenges AI brings to the traditional agency service model.
Via the Wall Street Journal, the Barclays’ analysts ruminated:
“We have been longstanding agency ‘bulls’…but we came away from all these meetings more bearish than before (…) While we still believe that agencies will adapt, survive and ultimately thrive (the questionable industry joke often told goes ‘we are cockroaches, not dinosaurs’), it will take time, money and good execution. We therefore see the current low growth persisting for longer than we initially thought.”
Not everyone was as bearish as Barclays.
The WSJ paraphrased a recent note from Bank of America analysts who said that “even though there’s an investor perception that AI increases the risk marketers will take agency tasks in-house, many marketers themselves seem to be playing down that possibility.”
Meanwhile, Goldman Sachs thinks differently than its competitors as the WSJ summarized a recent GS investor note:
“AI-driven improvements in personalization and targeting could significantly improve the return that advertisers achieve on their current spending, in turn perhaps unlocking larger ad budgets as advertisers redeploy those greater returns into other ad channels.”
Read more in The Wall Street Journal (June 30).
Prompt: What does a successful web publisher’s revenue strategy look like in 5 years?
Response from xAI’s Grok:
Predicting a successful web publisher’s revenue strategy for 2030 involves analyzing current trends, technological advancements, and shifts in consumer behavior, while acknowledging the dynamic nature of the digital publishing landscape.
Based on recent industry insights and patterns, a successful revenue strategy in 2030 will likely be diversified, adaptive, and heavily focused on audience engagement, data-driven personalization, and emerging technologies.
Below is a detailed outline of what such a strategy might look like:
1. Diversified Revenue Streams
To mitigate risks from over-reliance on any single income source, successful publishers will diversify their revenue streams, balancing direct and indirect monetization methods. According to industry reports, publishers are already pursuing an average of three to six revenue streams, and this trend is expected to intensify by 2030.
Key components include:
-
Subscriptions and Paywalls: Dynamic and flexible paywalls will remain a cornerstone, with an emphasis on personalized subscription models. Publishers like The New York Times have seen success with bundled subscriptions, where digital-only average revenue per user (ARPU) grew by 1.8% year-over-year in 2024, and bundled subscribers generated 44% higher monthly revenue than news-only subscribers. By 2030, publishers will likely offer tiered subscriptions that combine premium content, exclusive experiences (e.g., virtual events, podcasts), and value-added services like e-commerce discounts or educational resources.
-
Events and Experiences….
Retail media vulnerability
Even with a spicy headline – “Could AI Wipe Out a $200 Billion Ad Category?” – Next in Media’s Mike Shields caveated his story’s premise by admitting skepticism with the idea that we’ll all be handing over day-to-day tasks to robots.
Nevertheless, his journalistic “eye” was on the $200 billion-ish retail media ad market which could be vulnerable given AI’s evolution. Mr. Shields quotes eMarketer analyst Susan Marzano, “I think retailers are going to have to really work on loyalty program differentiation in order to stand out if agentic AI commoditizes the shopping experience.”
Trend: Making the MCP case
Claude AI Product Lead Scott White of LLM-maker Anthropic discussed the importance of Model Context Protocol – or MCP – in a Venture Beat interview published on Monday. Read it.
Anthropic originally developed the protocol which has received support from ad tech leaders such as LiveRamp CEO Scott Howe who said last month, MCP “creates a standardized way for models to access source data — without forcing everyone into a single ecosystem.”
From VentureBeat on Monday:
“A critical infrastructure development has been Anthropic’s Model Context Protocol (MCP), which White describes as ‘the USB-C for integrations.’ Rather than companies building separate connections to each data source or tool, MCP provides a standardized way for AI systems to access enterprise software, from Salesforce to internal knowledge repositories.”
“‘It’s really democratizing access to data,’ White said, noting that integrations built by one company can be shared and reused by others through the open-source protocol.”
”For organizations looking to implement AI agents, White recommends starting small and building incrementally. ‘Don’t try to build an entire agentic system from scratch,” he advised. ‘Build the component of it, make sure that component works, then build a next component.’”
More: “Introducing the Model Context Protocol” (November 2024) – Anthropic
More stuff
- Elon Musk’s xAI raises $10 billion in debt and equity as it steps up challenge to OpenAI (June 30) – CNBC
- Grammarly to acquire email startup Superhuman in AI platform push (July 1) – Reuters
- Podcast: “Michael Treff, CEO Code and Theory, Why Your AI Strategy Needs to Be More Than Tools & Efficiency” (June 30) – CMO Confidential on YouTube
- Senate strips AI provision from megabill (July 1) – The Hill
- DC’s AI Policy Collision, How Budget Reconciliation Became the Battleground for America’s AI Future (June 30) – Sustainable AI, a substack